The association of the international normalized ratio (INR) with the long-term clinical outcome of patients who undergo endarterectomy has not yet been studied. The present study therefore primarily aimed to evaluate the association of INR on admission with the 90-day mortality of critically ill patients who underwent endarterectomy during hospitalization. The Medical Information Mart for Intensive Care III database was queried for patients undergoing endarterectomy. The 90-day mortality of patients was selected as a primary endpoint. Receiver-operating characteristic (ROC) curves were plotted to present the accuracy of predictions. Kaplan-Meier curves and multivariate Cox regression analysis were performed to analyse associations. Propensity score matching (PSM) was also conducted to reduce confounding bias. A total of 230 patients were included, with 36 90-day non-survivors. Patients with a high INR (≥1.5) on admission exhibited a higher 90-day mortality than those with a low INR (<1.5; 29.09 vs. 11.43%; P=0.003). The ROC area under the curve value was 0.687 [95% confidence interval (CI), 0.571–0.780]. Kaplan-Meier plots identified divergence in survival between patients with different INR levels (log-rank test, P=0.0013). The results of the multivariate Cox regression analysis indicated that a high INR level was significantly associated with 90-day mortality (hazard ratio, 2.19; 95% CI, 1.08–4.45; P=0.0305). Analysis of the PSM cohort presented similar results. In conclusion, the INR levels of critically ill patients who undergo endarterectomy may be used to stratify their risk of 90-day mortality.